Human salivary mucins protect the oral tissues by providing a physical barrier to environmental agents, possessing viscoelastic properties essential for lubrication and participating in the modulation of the oral flora. These properties are dependent, in large part, on the mucin's O- linked oligosaccharides and the peptide moieties surrounding the glycopeptide linkage. Several central questions regarding the patterns of O-glycosylation remain to be understood for the prediction of the influence of O-glycosylation patterns on the mucin's biological function(s). First, a consensus sequence akin to the Asn-Xxx-Thr/Ser motif for N-glycosylation has not been identified and hence the factor(s) that determine which Ser and Thr residues are O-glycosylated are not known. Second, factors that determine the final structure of the O- linked units and their distribution along the peptide backbone are not understood. Our hypothesis is that the patterns of MG2 O-glycosylation are important determinants in biological activity. Importantly, these patterns are influenced by the conformation adopted by O-linked units and adjacent or flanking peptide sequences. For example, mucin's interaction with certain bacteria (e.g. viridans streptococci and Actinomyces viscosus) involves stereochemical recognition between mucin's O-linked oligosaccharides and bacterial adhesins. Our long range goal is to obtain a detailed understanding of human salivary mucin O- glycosylation patterns for the eventual production by design and mimicry of bioactive mucin analogs to be used as saliva substitutes. To achieve this goal, we will use MG2 as a model since its structure has been well characterized in our laboratory. Our approach to this problem is biphasic in nature. In the first phase, we will synthesize MG2 glycopeptides present in the tandem repeat with varying levels of O- linked units. The bioactivity of these glycopeptides will be compared to native MG2 to assess the role of glycosylation patterns on biological function. In the later phase, a complete understanding of the conformational dynamics of these bioactive glycopeptides will be carried out. The data obtained will then be used to design mucin peptides (e.g. glycomimetics) that mimic the conformation and biological properties of bioactive glycopeptides.